Closed eduherminio closed 3 days ago
Simple impl, single value vs next impl
Test | eval/eval/bad-bishop-penalty-retune-1 Elo | -8.05 +- 5.65 (95%) SPRT | 8.0+0.08s Threads=1 Hash=32MB LLR | -2.25 (-2.25, 2.89) [0.00, 3.00] Games | 6866: +1902 -2061 =2903 Penta | [229, 835, 1417, 770, 182] https://openbench.lynx-chess.com/test/734/
Convert to array and add tuned values See #1022
Use pawn diff
Use improved pawn diff
Test | eval/bad-bishop-penalty-retune-3 Elo | -1.75 +- 3.12 (95%) SPRT | 8.0+0.08s Threads=1 Hash=32MB LLR | -2.27 (-2.25, 2.89) [0.00, 3.00] Games | 22690: +6337 -6451 =9902 Penta | [661, 2802, 4531, 2692, 659] https://openbench.lynx-chess.com/test/735/
Add blocked pawns?
Simple impl, single value vs next impl
Convert to array and add tuned values See #1022
Use pawn diff
Use improved pawn diff